Temporal and Developmental Requirements for the Prader–Willi Imprinting Center

Temporal and developmental requirements for the Prader–Willi imprinting center

Amanda J. DuBose1,2, Emily Y. Smith2, Karen A. Johnstone3, and James L. Resnick4

Department of Molecular Genetics and Microbiology, University of Florida College of Medicine, Gainesville, FL 32610

Edited by Arthur L. Beaudet, Baylor College of Medicine, Houston, TX, and approved January 17, 2012 (received for review September 13, 2011) Imprinted gene expression associated with Prader–Willi syndrome neurogenesis does reduce Snrpn and snoRNA expression, but (PWS) and Angelman syndrome (AS) is controlled by two imprint- this effect is likely a result of the removal of the Snrpn major ing centers (ICs), the PWS-IC and the AS-IC. The PWS-IC operates in promoter. These pups appear to bypass the early failure to thrive cis to activate transcription of genes that are expressed exclusively seen in other PWS models but still exhibit a reduction in post- from the paternal allele. We have created a conditional allele of weaning weight gain. These results conﬁrm that the PWS-IC is the PWS-IC to investigate its developmental activity. Deletion of necessary postzygotically for paternal gene expression but also the paternal PWS-IC in the embryo before implantation abolishes demonstrate that it is not required later in development for expression of the paternal-only genes in the neonatal brain. Sur- maintenance of the paternal epigenotype. prisingly, deletion of the PWS-IC in early brain progenitors does not affect the subsequent imprinted status of PWS/AS genes in the Results newborn brain. These results indicate that the PWS-IC functions to Generation of PWS-IC Conditional Deletion Allele. We previously protect the paternal epigenotype at the epiblast stage of develop- demonstrated that the entire murine PWS-IC lies within a 6-kb ment but is dispensable thereafter. region located between −3.7 kb and +2.3 kb with reference to Snrpn exon 1. Furthermore, we generated a conditional PWS-IC rader–Willi syndrome (PWS) and Angelman syndrome (AS) allele by ﬂanking this region with loxP sites (5). The structure of are genetic disorders resulting from the loss of expression this conditional allele is shown in Fig. S1.

P GENETICS of a cluster of imprinted genes at chromosome 15q11-q13. Individuals with PWS lack paternal gene expression from this Deletion of PWS-IC Before Implantation Results in Reduced Birth imprinted region whereas individuals with AS lack maternal gene Weight and Neonatal Lethality. To investigate the effects of early embryonic deletion of the PWS-IC, we analyzed offspring from expression (1). In contrast to imprinted domains controlled by fl a mating of males heterozygous for the PWS-IC ox6kb allele with a single imprinting center, the PWS/AS region is regulated by Cmv-cre a bipartite imprinting center composed of the AS imprinting transgenic (Tg) Cmv-cre (Tg ) hemizygous females. The center (AS-IC) and the PWS-IC (2). Imprinting in the region Cmv-cre transgene expresses Cre ubiquitously before implantation (12). Similar to paternal inheritance of a PWS-IC deletion results from interplay between the AS-IC and the PWS-IC. +/flox6kb Cmv-cre Substantial evidence suggests that the PWS-IC is a positive-act- (4, 5), PWS-IC Tg pups were visibly smaller and fi ing element that stimulates the expression of genes in the 2-Mb had signi cantly lower birth weights than littermates (Fig. 1 A – PWS/AS region. The AS-IC is thought to act in the female and B). Cmv-cre mediated deletion of the PWS-IC before im- germline to inactivate the PWS-IC on the future maternal allele plantation caused 100% postnatal lethality with no pups surviv- (3). Thus, a functional PWS-IC is necessary for the expression of ing beyond postnatal day (P) 2 (Fig. 1C). Southern blot analysis paternal-only genes and an intact AS-IC is necessary to silence of DNA isolated from several P1 brains suggests that Cmv-cre these genes on the maternal chromosome. Using a targeted mediated deletion is extensive (Fig. S2). As expected, mice with deletion of a 35-kb region spanning Snrpn exons 1 to 6, we a somatic deletion of the PWS-IC on the maternal chromosome were normal, as were mice inheriting a maternal or paternal previously reported that the location and function of the PWS-IC flox6kb is conserved in mice (4). More recently, we found that a 6-kb PWS-IC allele in the absence of Cre expression (5). deletion surrounding Snrpn exon 1 yields an identical phenotype to the 35-kb deletion, demonstrating that all the functional ele- Paternal Gene Expression Requires Presence of PWS-IC in Preim- ments of the PWS-IC are within this 6-kb region (5). plantation Embryo. To determine the extent of the imprinting The PWS-IC lies within a region of allele-specific DNA meth- defect resulting from preimplantation deletion of the PWS-IC, we examined gene expression in RNA isolated from the brains of ylation. The silent maternal allele is hypermethylated whereas +/flox6kb Cmv-cre the paternal allele is hypomethylated. These DNA methylation P1 PWS-IC Tg pups. Deletion of the PWS-IC be- imprints are erased and reestablished in the germline during ga- fore implantation affected the expression of all genes in the re- metogenesis (6–10). When the PWS-IC acts in somatic tissue to gion. Snrpn, Frat3, Ndn, Magel2, Mkrn3, Snord116,andSnord115 regulate allele-specific gene expression is unknown. Bielinska et al. (11) found that the PWS-IC is required postzygotically in both humans and mice. An individual with minor clinical symptoms of Author contributions: A.J.D., E.Y.S., K.A.J., and J.L.R. designed research; A.J.D., E.Y.S., K.A.J., and J.L.R. performed research; A.J.D., E.Y.S., and K.A.J. contributed new re- PWS was shown to be mosaic for a PWS-IC deletion, suggesting agents/analytic tools; A.J.D., E.Y.S., K.A.J., and J.L.R. analyzed data; and A.J.D., E.Y.S., that the PWS-IC is necessary to maintain paternal gene expression K.A.J., and J.L.R. wrote the paper. postzygotically. Mice chimeric for a PWS-IC deletion on the pa- The authors declare no conflict of interest. ternal chromosome exhibited promoter hypermethylation and loss This article is a PNAS Direct Submission. of expression of Ndn and Mkrn3, supporting the interpretation 1Present address: National Human Genome Research Institute, National Institutes of that the PWS-IC is required postzygotically (11). Health, Bethesda, MD 20892. We have created a conditional allele of the PWS-IC to further 2A.J.D. and E.Y.S. contributed equally to this work. investigate the temporal and spatial functions of this element. As 3Present address: Institute of Biomedical and Clinical Science, Peninsula Medical School, predicted from previous work, we find that the PWS-IC is nec- University of Exeter, Exeter EX2 5DW, United Kingdom. essary before implantation. Unexpectedly, deletion of the PWS- 4To whom correspondence should be addressed. E-mail: [email protected]fl.edu. IC early in neurogenesis does not affect subsequent expression of This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. imprinted genes at the locus in newborn brain. Deletion early in 1073/pnas.1115057109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1115057109 PNAS Early Edition | 1of5 Downloaded by guest on September 26, 2021 Fig. 1. PWS-IC+/flox6kb TgCmv-cre pups are small and have reduced survival compared with littermates of other genotypes. (A) The PWS-IC+/flox6kb TgCmv-cre flox neonate on the left is smaller than the WT littermate on the right (P1). (B) PWS-IC+/ 6kb TgCmv-cre mice weigh significantly less than mice of other genotypes flox at birth (P = 0.0007; n = 11 and n = 59, respectively). (C) PWS-IC+/ 6kb TgCmv-cre mice (circles, n = 11) exhibit postnatal lethality with no survivors beyond P2 compared with other genotypes (squares, n = 46).

fl expression was not detected in PWS-IC+/ ox6kb TgCmv-cre newborn heterozygous for B6.cast.c7 (congenic for Mus musculus casta- fl mice (Fig. 2). Gene expression patterns in PWS-IC+/ ox6kb pups neus across a portion of chromosome 7) were crossed with PWS- fl that did not inherit TgCmv-cre were indistinguishable from WT. IC+/ ox6kb males, and the progeny were analyzed at 12 wk of age. As shown for other mutations that down-regulate Snrpn–Ube3a- Deletion of PWS-IC in Neuronal Precursors Does Not Result in an as expression, we also found an increase in paternal contribution fl Imprinting Defect. Cmv-cre–mediated deletion of the PWS-IC of Ube3a in PWS-IC+/ ox6kb TgNes-cre mice (Fig. S3). results in a complete imprinting defect and thereby demonstrates that the PWS-IC must be intact in the preimplantation embryo to Paternal Epigenotype at Ndn and Mkrn3 Is Stable Postzygotically. maintain the paternal pattern of gene expression during sub- The Ndn and Mkrn3 promoters exhibit differential DNA meth- sequent embryonic development. We next sought to establish ylation with hypermethylation of the maternal allele (22, 23). whether the PWS-IC is required to maintain the imprint at later Digestion with methyl-sensitive restriction endonucleases was flox6kb times in development. PWS-IC males were mated to used to analyze the DNA methylation status of Mkrn3 and Ndn fl fl transgenic Nes-cre females to mediate PWS-IC deletion in in both PWS-IC+/ ox6kb TgCmv-cre and PWS-IC+/ ox6kb TgNes-cre neuronal precursors and glia by embryonic day (E) 10.5 with mice. Germline transmission of the 6-kb PWS-IC deletion led to widespread deletion by E12.5 (13, 14). The paternally expressed the paternal allele adopting a maternal DNA methylation pat- genes Ndn, Frat3, Magel2, and Mkrn3 were unaltered in the tern at both Mkrn3 and Ndn (5), as did deletion with Cmv-cre. +/flox6kb Nes-cre brains of PWS-IC Tg newborns (Fig. 3). Normal Nes-cre–mediated deletion, which does not cause an imprinting expression of these genes indicates that the PWS-IC is not nec- defect, resulted in a WT DNA methylation pattern at both loci essary to maintain paternal gene expression in brain precursor (Fig. 5). Thus, the DNA methylation imprints at Ndn and Mkrn3 cells beyond E12.5. are stable in the brain even in the absence of the PWS-IC. The downstream cluster of genes, Snrpn, several snoRNAs, and Ube3a-as, are likely processed from a common 1-Mb tran- Deletion of PWS-IC in Neuronal Precursors Results in Postweaning script (15–17). Deletion of the PWS-IC also removes the major Growth Deficiency. The reduction in Snrpn–Ube3a-as expression promoter for this transcript. Snrpn expression in the brains of P1 fl in the nervous system led us to investigate growth traits of PWS- PWS-IC+/ ox6kb TgNes-cre mice is reduced by almost 70%, whereas the snoRNAs, Snord116 and Snord115, are reduced by approximately half (Fig. 3). Nes-cre–mediated deletion of the PWS-IC in P1 brain was extensive (Fig. S2), indicating that the observed residual gene expression was not the result of in- complete PWS-IC deletion. We investigated whether deletion of the major promoter for this transcript contributed to the reduction in expression of Snrpn and the snoRNAs. Several distant upstream exons (U exons) for Snrpn have previously been found to contribute to transcription of the locus (15, 18). We tested whether the residual expression of Snrpn and the snoRNAs could originate fl from the Snrpn U exons in PWS-IC+/ ox6kb TgNes-cre animals. Transcripts originating at Snrpn upstream exon U1 and extending fl to Snrpn exon 3 were detectable in the brains of PWS-IC+/ ox6kb TgNes-cre mice but not in mice inheriting the same deletion via the fl germline, nor in PWS-IC+/ ox6kb TgCmv-cre mice (Fig. 4). These results confirm and extend previous findings that the U exons contribute to the Snrpn transcription unit and also demonstrate that U exon use requires the presence of the PWS-IC before implantation (18). Paternal expression of the Snrpn transcription unit is widely thought to contribute to silencing of Ube3a on the paternal allele +/flox6kb Cmv-cre – Fig. 2. Expression of PWS genes in PWS-IC Tg P1 brains. Ex- via antisense transcription (19 21). Elimination of the major pression of the indicated genes was determined by either Northern blot or Snrpn promoter prompted us to investigate allelic expression of RT-PCR. A minus sign to the left of RT-PCR analyses indicates control samples Ube3a. Females heterozygous for the Nes-cre transgene and in which reverse transcriptase was omitted during cDNA synthesis.

2of5 | www.pnas.org/cgi/doi/10.1073/pnas.1115057109 DuBose et al. Downloaded by guest on September 26, 2021 Fig. 4. Snrpn upstream exon use in the brain. Snrpn upstream exon transcripts were detected by RT-PCR using primers in upstream exon U1 and exon 3. (A) Expression from Snrpn upstream exon 1 is not detected following germline transmission of the PWS-IC deletion. (B) Snrpn upstream exon transcripts following Cmv-cre– or Nes-cre–mediated deletion.

of a previously described 35-kb PWS-IC deletion and indicates that the entire murine PWS-IC is included within the boundaries of the 6-kb deletion. Here, we have characterized the con- sequences of conditional deletion of this minimal PWS-IC. Evidence from both a mosaic individual with some features of PWS and chimeric mice suggests that the PWS-IC is necessary postzygotically for paternal PWS gene expression and to avoid a maternal methylation imprint (11). Both the mosaic individual

and chimeric mice transmitted the deletion, implying that the GENETICS PWS-IC may not be necessary after separation of the germline from soma. Cmv-cre–mediated deletion of the PWS-IC before implantation results in a gene expression pattern indistinguishable from germline transmission of a PWS-IC deletion, thus indicating that the PWS-IC is necessary before or during implantation. In contrast, Nes-cre–mediated deletion of the PWS-IC Fig. 3. Expression of upstream PWS genes in P1 brains is preserved whereas flox expression of the downstream cluster of genes is reduced in PWS-IC+/ 6kb in brain progenitors, where extensive deletion occurs by E12.5, TgNes-cre mice. (A) Gene expression assays were as indicated in Fig. 2. (B)Ex- does not lead to an imprinting defect. In this case, expression of pression levels of Snrpn, Ndn, Snord116, and Snord115 were quantified by Mkrn3, Magel2, Frat3, and Ndn in newborn brain is unaffected by PhosphorImager analysis of the Northern blots. The average of the three the absence of the PWS-IC. Together, these results establish that flox control lanes ± SE was compared with the average of the three PWS-IC+/ 6kb the presence of the PWS-IC is required in the period before im- Nes-cre Tg lanes and normalized to β-actin (Snrpn and Ndn) or 5.8S RNA plantation through formation of early neural precursors. This Snord115 116 P ( and ) exposures on the same blot. values for reduction in period is characterized by dynamic epigenetic changes in somatic Snrpn, Snord116, and Snord115 expression levels are 0.018, 0.017, and 0.142, respectively. Ndn level was not significantly altered (P = 0.2, Student t test). tissues. Fertilization is followed by DNA demethylation of the paternal and maternal genomes. The embryonic genome is then remethylated at the epiblast stage early in gastrulation (reviewed fl IC+/ ox6kb TgNes-cre deletion mice. Pups inheriting a PWS- in ref. 24). This stage of genomic DNA methylation is likely fl IC ox6kb allele paternally and the Nes-cre transgene maternally a critical period for PWS-IC function. Our data are consistent with did not exhibit a statistically significant difference in birth weight. a model in which the PWS-IC prevents acquisition of a maternal fl PWS-IC+/ ox6kb TgNes-cre pups regularly exhibited milk spots and imprint on the paternal allele during this developmental stage. fl lacked the pallor characteristic of PWS-IC+/ ox6kb TgCmv-cre The finding that the paternal epigenotype can be maintained newborns or pups inheriting a germline deletion of the PWS-IC throughout much of neurogenesis in the absence of the PWS-IC fl (5). PWS-IC+/ ox6kb TgNes-cre pups were born with the expected was unexpected. The contribution of epigenetic modifications to Mendelian frequency (14 of 52; 26.9%) and did not show a re- neural development is poorly understood, although several duction in postnatal survival compared with littermates of other reports illuminate the importance of this form of gene regula- fl genotypes. However, PWS-IC+/ ox6kb TgNes-cre mice did exhibit tion. The human brain exhibits region specific DNA methylation a growth retardation that became increasingly apparent during patterns, suggesting that DNA methylation contributes to neural the postweaning period (Fig. 6). By 3 wk of age, female PWS- development (25). A role for epigenetic regulation in brain de- fl IC+/ ox6kb TgNes-cre mice displayed a significant growth deficiency velopment is further suggested by mutations in DNA methyl- (8.32 ± 0.48 vs. 10.1 ± 0.32; P = 0.022) and, although their transferases (reviewed in ref. 26). Our results indicate that the growth curve was similar to littermates, they remained growth- epigenotype of paternally expressed genes at the PWS/AS locus restricted (Fig. 6A). Males of the same genotype also exhibited is dependent upon the PWS-IC during early embryonic de- a growth deficiency that reached statistical significance by 4 wk of velopment. However, by the time Nes-cre becomes active in age (15.1 ± 0.52 vs. 18.1 ± 0.36; P = 0.012; Fig. 6B). neural cells at E12.5, maintenance of the paternal epigenotype no longer requires the PWS-IC. Discussion Previous efforts to conditionally inactivate elements regulating We recently demonstrated that paternal transmission of a 6-kb imprinted gene expression during embryonic development have deletion spanning Snrpn exon 1 leads to low birth weight, neo- focused on the H19/Igf2 and Rasgrf1 loci. H19 is normally silenced natal lethality, and complete loss of PWS gene expression in on the paternal allele whereas Igf2 is normally silenced on the newborn brain (5). This phenotype is indistinguishable from that maternal chromosome. Germline transmission of a paternal IC

DuBose et al. PNAS Early Edition | 3of5 Downloaded by guest on September 26, 2021 Fig. 6. Postweaning growth of PWS-IC+/flox6kb TgNes-cre mice and littermates. Mice were weighed weekly after weaning. Weight curves for female (A)and flox male (B) PWS-IC+/ 6kb TgNes-cre mice and their littermates (n =6andn =32for flox Fig. 5. DNA methylation analysis of the Ndn and Mkrn3 differentially females and n =8andn = 20 for males, PWS-IC+/ 6kb TgNes-cre and other +/+ +/Δ35kb +/Δ6kb flox methylated regions. DNA from brains of WT ( ), PWS-IC , PWS-IC , mice, respectively). The gray line shows weight gain for PWS-IC+/ 6kb TgNes-cre +/flox6kb Cmv-cre +/flox6kb Nes-cre PWS-IC Tg , and PWS-IC Tg pups was digested with mice and the black line shows the weight gain for the other genotypes. Error the indicated restriction endonucleases, Southern blotted, and probed. (A) bars indicate SE. The upper part shows a restriction map of the Ndn locus. The DNA methylation sensitive endonuclease SacII (S) is located between two HindIII (H) sites. The Ndn probe detects a 3.3-kb fragment (Me) resulting from DNA methyl- the PWS-IC or a deletion spanning from Snrpn to Ube3a (18, 33, fl ation at the SacII site and a 1.8-kb fragment obtained from unmethylated 34). In the present PWS-IC ox model, the Snrpn U exons drive B Mkrn3 DNA (UnMe). ( ) The upper part shows a restriction map of the lo- expression of all of the downstream cluster gene products in- cus. A site recognized by the DNA methylation sensitive NotI (N) endonuclease is located between HindIII and SpeI (Sp) sites at the Mkrn3 locus. The dicating that the upstream exons are not exclusively dedicated to probe detects methylated and unmethylated fragments of 6.8 and 5.0 kb, a particular processing variant. Conditional deletion of the PWS- respectively. IC with Nes-cre suggests that even modest changes in the level of the Snrpn transcription unit can significantly affect body weight. Previous studies have also shown that components of the large deletion at this locus results in expression of the normally silent Snrpn to Ube3a-as transcription unit affect body weight (18, 21, paternal H19 allele (27, 28). Conditional deletion of the paternal 34–36). Inheritance of a Snrpn promoter that is not maternally IC in zygotes also leads to biallelic H19 expression. However, H19 silenced causes an increase in downstream transcription products expression remains exclusively maternal after paternal IC de- and a significant increase in body weight (21). Unlike previous letion in terminally differentiated skeletal, cardiac, or liver cells, PWS-IC mutants, conditional deletion in brain progenitors of the suggesting that the imprinted status of H19 becomes IC-in- PWS-IC by Nes-cre does not result in neonatal lethality or overt dependent sometime after implantation. The Rasgrf1 locus is signs of neonatal failure to thrive. These results suggest that the paternally expressed. Acquisition of DNA methylation at a dif- small size of these mutants is not a result of neonatal failure to ferentially methylated domain in the male germline is dependent thrive but rather a reduction in the postweaning activity of the upon a set of repeats located 30 kb upstream of the transcription snoRNAs, as growth retardation is characteristic of other mouse start site. A conditional deletion strategy targeting these repeats models featuring reduced Snord116 expression (34–36). demonstrated that, although the repeats are necessary to maintain differential methylation after fertilization, they are no longer Materials and Methods necessary after the epiblast stage (29). Unlike these loci, the DNA Animals. All animal procedures were previously approved by the University of methylation imprint is maternally acquired at the PWS/AS locus. Florida Institutional Animal Care and Use Committee. The PWS-ICflox6kb allele The paternal PWS-IC is necessary to establish expression of was created as previously described (5). B6.C-Tg (Cmv-cre)1Cgn/J (stock no. paternal genes but is dispensable for paternal gene expression 006054) and B6.C-Tg(Nes-cre)1Kln/J (stock no. 006054) were obtained from by the time neural precursors are formed. The Rasgrf1 study the Jackson Laboratory. The Nes-cre females were used after 13 to 15 gen- erations of backcross to C57BL/6. The B6.cast.c7 line has been previously demonstrates stability of the paternal methylated state of the described (20). Animal weights are expressed as averages ± SE. Statistical differentially methylated domain after implantation, whereas the significance was determined by unpaired Student t test. H19/Igf2 studies establish that the paternal IC is dispensable for continued paternal H19 silencing. Gene Expression Analysis. Northern Blot analysis and RT-PCR were performed Snrpn, Snord116, Snord115, and Ube3a-as RNAs are gener- as previously described (5). RT-PCR for the Snrpn upstream exon use was ated from primary transcripts with a common initiation site (15– performed as described previously (18). Northern blot band intensity was 17, 30–32). Expression of these genes is reduced following Nes- digitized on the Storm 860 PhosphorImager (GE Healthcare) and analyzed cre-mediated deletion, most likely because of the absence of the with ImageQuant TL software. major promoter at Snrpn exon 1 rather than epigenetic modifications of the locus. As previously suggested, residual tran- DNA Methylation Analysis. DNA was isolated from newborn brain, digested scription of these genes is likely to initiate at several upstream with the indicated restriction endonucleases, and analyzed by Southern blot as previously described (5). Probe details are available on request. promoters that splice into Snrpn exon 2 or into acceptor sites located further downstream (18). ACKNOWLEDGMENTS. The authors thank Theresa Strong, Rob Nicholls, Tom Conditional deletion of the PWS-IC with Nes-cre results in Yang, and Brian Harfe for helpful comments on the manuscript; Deb Morse adults that are smaller than WT littermates. The only significant and Ryan Hallett for technical assistance; Ryan Fiske of the University of alteration in PWS gene expression is a reduction in Snrpn and the Florida Mouse Models Core; and Susan D’Costa, Carlos Sulsona, and Justin Bickford. This work was supported by grants from the Foundation for snoRNAs, and an increase in paternal expression of Ube3a. In this Prader–Willi Research and National Institute of Child Health and Human regard, these animals resemble other models lacking expression of Development Grant R01 HD 037872. E.Y.S. was supported by a University the Snrpn transcription unit as a result of a germline deletion of of Florida Alumni Fellowship.

1. Nicholls RD, Knepper JL (2001) Genome organization, function, and imprinting in 2. Buiting K (2010) Prader-Willi syndrome and Angelman syndrome. Am J Med Genet C Prader-Willi and Angelman syndromes. Annu Rev Genomics Hum Genet 2:153–175. Semin Med Genet 154C:365–376.

4of5 | www.pnas.org/cgi/doi/10.1073/pnas.1115057109 DuBose et al. Downloaded by guest on September 26, 2021 3. Brannan CI, Bartolomei MS (1999) Mechanisms of genomic imprinting. Curr Opin 21. Johnstone KA, et al. (2006) A human imprinting centre demonstrates conserved ac- Genet Dev 9:164–170. quisition but diverged maintenance of imprinting in a mouse model for Angelman 4. Yang T, et al. (1998) A mouse model for Prader-Willi syndrome imprinting-centre syndrome imprinting defects. Hum Mol Genet 15:393–404. Nat Genet – mutations. 19:25 31. 22. Hanel ML, Wevrick R (2001) Establishment and maintenance of DNA methylation fi 5. Dubose AJ, Smith EY, Yang TP, Johnstone KA, Resnick JL (2011) A new deletion re nes patterns in mouse Ndn: implications for maintenance of imprinting in target genes of the boundaries of the murine Prader-Willi syndrome imprinting center. Hum Mol the imprinting center. Mol Cell Biol 21:2384–2392. Genet 20:3461–3466. 23. Jong MT, et al. (1999) A novel imprinted gene, encoding a RING zinc-finger protein, 6. Glenn CC, Porter KA, Jong MT, Nicholls RD, Driscoll DJ (1993) Functional imprinting and overlapping antisense transcript in the Prader-Willi syndrome critical region. Hum and epigenetic modification of the human SNRPN gene. Hum Mol Genet 2: Mol Genet – 2001–2005. 8:783 793. 7. Shemer R, Birger Y, Riggs AD, Razin A (1997) Structure of the imprinted mouse Snrpn 24. Sasaki H, Matsui Y (2008) Epigenetic events in mammalian germ-cell development: Nat Rev Genet – gene and establishment of its parental-specific methylation pattern. Proc Natl Acad Reprogramming and beyond. 9:129 140. Nature – Sci USA 94:10267–10272. 25. Dulac C (2010) Brain function and chromatin plasticity. 465:728 735. 8. Zeschnigk M, et al. (1997) Imprinted segments in the human genome: Different DNA 26. Feng J, et al. (2010) Dnmt1 and Dnmt3a maintain DNA methylation and regulate methylation patterns in the Prader-Willi/Angelman syndrome region as determined synaptic function in adult forebrain neurons. Nat Neurosci 13:423–430. by the genomic sequencing method. Hum Mol Genet 6:387–395. 27. Srivastava M, et al. (2000) H19 and Igf2 monoallelic expression is regulated in two 9. Hajkova P, et al. (2002) Epigenetic reprogramming in mouse primordial germ cells. distinct ways by a shared cis acting regulatory region upstream of H19. Genes Dev 14: Mech Dev 117:15–23. 1186–1195. 10. Sutcliffe JS, et al. (1994) Deletions of a differentially methylated CpG island at the 28. Thorvaldsen JL, Fedoriw AM, Nguyen S, Bartolomei MS (2006) Developmental profile fi Nat Genet – SNRPN gene de ne a putative imprinting control region. 8:52 58. of H19 differentially methylated domain (DMD) deletion alleles reveals multiple roles De novo SNRPN 11. Bielinska B, et al. (2000) deletions of exon 1 in early human and of the DMD in regulating allelic expression and DNA methylation at the imprinted mouse embryos result in a paternal to maternal imprint switch. Nat Genet 25:74–78. H19/Igf2 locus. Mol Cell Biol 26:1245–1258. 12. Schwenk F, Baron U, Rajewsky K (1995) A cre-transgenic mouse strain for the ubiq- 29. Holmes R, Chang Y, Soloway PD (2006) Timing and sequence requirements defined uitous deletion of loxP-flanked gene segments including deletion in germ cells. Nu- for embryonic maintenance of imprinted DNA methylation at Rasgrf1. Mol Cell Biol cleic Acids Res 23:5080–5081. – 13. Tronche F, et al. (1999) Disruption of the glucocorticoid receptor gene in the nervous 26:9564 9570. fi fi system results in reduced anxiety. Nat Genet 23:99–103. 30. Cavaillé J, et al. (2000) Identi cation of brain-speci c and imprinted small nucleolar Proc Natl Acad Sci USA 14. Graus-Porta D, et al. (2001) Beta1-class integrins regulate the development of laminae RNA genes exhibiting an unusual genomic organization. 97: and folia in the cerebral and cerebellar cortex. Neuron 31:367–379. 14311–14316. 15. Landers M, et al. (2004) Regulation of the large (∼1000 kb) imprinted murine Ube3a 31. de los Santos T, Schweizer J, Rees CA, Francke U (2000) Small evolutionarily conserved antisense transcript by alternative exons upstream of Snurf. Snrpn. Nucleic Acids Res RNA, resembling C/D box small nucleolar RNA, is transcribed from PWCR1, a novel 32:3480–3492. imprinted gene in the Prader-Willi deletion region, which Is highly expressed in brain. 16. Le Meur E, et al. (2005) Dynamic developmental regulation of the large non-coding Am J Hum Genet 67:1067–1082.

Dev Biol GENETICS RNA associated with the mouse 7C imprinted chromosomal region. 286: 32. Wevrick R, Francke U (1997) An imprinted mouse transcript homologous to the hu- – 587 600. man imprinted in Prader-Willi syndrome (IPW) gene. Hum Mol Genet 6:325–332. NURF-SNRPN 17. Runte M, et al. (2001) The IC-S transcript serves as a host for multiple 33. Chamberlain SJ, et al. (2004) Evidence for genetic modiﬁers of postnatal lethality in small nucleolar RNA species and as an antisense RNA for UBE3A. Hum Mol Genet 10: PWS-IC deletion mice. Hum Mol Genet 13:2971–2977. 2687–2700. 34. Tsai TF, Jiang YH, Bressler J, Armstrong D, Beaudet AL (1999) Paternal deletion from 18. Bressler J, et al. (2001) The SNRPN promoter is not required for genomic imprinting of Snrpn to Ube3a in the mouse causes hypotonia, growth retardation and partial le- the Prader-Willi/Angelman domain in mice. Nat Genet 28:232–240. Hum 19. Rougeulle C, Cardoso C, Fontés M, Colleaux L, Lalande M (1998) An imprinted anti- thality and provides evidence for a gene contributing to Prader-Willi syndrome. Mol Genet – sense RNA overlaps UBE3A and a second maternally expressed transcript. Nat Genet 8:1357 1364. 19:15–16. 35. Ding F, et al. (2008) SnoRNA Snord116 (Pwcr1/MBII-85) deletion causes growth de- 20. Chamberlain SJ, Brannan CI (2001) The Prader-Willi syndrome imprinting center ac- ﬁciency and hyperphagia in mice. PLoS ONE 3:e1709. tivates the paternally expressed murine Ube3a antisense transcript but represses 36. Skryabin BV, et al. (2007) Deletion of the MBII-85 snoRNA gene cluster in mice results paternal Ube3a. Genomics 73:316–322. in postnatal growth retardation. PLoS Genet 3:e235.

DuBose et al. PNAS Early Edition | 5of5 Downloaded by guest on September 26, 2021